We propose new research to investigate unintended pregnancies and their social context. Our approach will provide new estimates of the prevalence of unintended pregnancy, new insights into the social context producing unintended pregnancies, and new measurement strategies that can be adapted by major national studies of unintended pregnancy and related topics. To advance our understanding of the processes leading to unintended pregnancy among young women, this project has four aims: (1) Collect new detailed, dynamic measures of unintended pregnancy;(2) Collect new detailed, dynamic measures of behavioral, attitudinal, and community context aspects of relationships, contraceptive use, pregnancy, and activities that compete with childbearing;(3) Provide alternative estimates of the prevalence of unintended pregnancy based on new, prospective measurement strategies and directly compare these new prevalence estimates to those based on national studies;and (4) Determine which behavioral, attitudinal, and contextual aspects of relationships, contraceptive use, and activities that compete with childbearing increase unintended pregnancy rates during the transition to adulthood. To meet these aims, we propose a five-year study that begins with ethnography, cognitive interviewing, and a pilot study - all coordinated with an advisory group of national data collection leaders - to finalize the data collection instruments we will use. Next we will conduct face-to-face interviews with 1250 18 and 19-year-old women in one county in Michigan, along with a thirty month journal follow-up study. Over the thirty month period, these women will provide weekly updates about behavioral and attitudinal aspects of their relationships, contraceptive use, pregnancies, and competing activities via Internet and Interactive Voice Response (telephone) technologies. We will also conduct semi-structured follow-up interviews with approximately 10 percent of the women who experience pregnancies and a control group of young women who do not become pregnant. At the conclusion of the journal follow-up period, women in the study will complete the individual interview used for Cycle 7 of the National Survey of Family Growth (NSFG). Adjusting for potential testing-response interactions, the NSFG-style measurement will provide a comparison between the weekly journal-based estimates and the national estimates of unintended pregnancy and related factors. Our analyses of the journal measures will feature state-of-the-art dynamic event history models using the fully detailed timing information to document the interconnections among behavioral, attitudinal, and contextual aspects of relationships, contraceptive use, activities that compete with childbearing, and unintended pregnancy. The results will produce essential new insights into the processes surrounding this high priority public health and public policy concern.